EP1609063B1 - Zustandswiederherstellung und failover intelligenter netzwerkadapter - Google Patents

Zustandswiederherstellung und failover intelligenter netzwerkadapter Download PDF

Info

Publication number
EP1609063B1
EP1609063B1 EP03816813A EP03816813A EP1609063B1 EP 1609063 B1 EP1609063 B1 EP 1609063B1 EP 03816813 A EP03816813 A EP 03816813A EP 03816813 A EP03816813 A EP 03816813A EP 1609063 B1 EP1609063 B1 EP 1609063B1
Authority
EP
European Patent Office
Prior art keywords
network
host
offload engine
adapter
state
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Lifetime
Application number
EP03816813A
Other languages
English (en)
French (fr)
Other versions
EP1609063A4 (de
EP1609063A1 (de
Inventor
Douglas Freimuth
Ronald Mraz
Erich Nahum
Prashant Pradhan
Sambit Sahu
John Tracey
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
International Business Machines Corp
Original Assignee
International Business Machines Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by International Business Machines Corp filed Critical International Business Machines Corp
Publication of EP1609063A1 publication Critical patent/EP1609063A1/de
Publication of EP1609063A4 publication Critical patent/EP1609063A4/de
Application granted granted Critical
Publication of EP1609063B1 publication Critical patent/EP1609063B1/de
Anticipated expiration legal-status Critical
Expired - Lifetime legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
    • H04L69/16—Implementation or adaptation of Internet protocol [IP], of transmission control protocol [TCP] or of user datagram protocol [UDP]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
    • H04B1/74—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission for increasing reliability, e.g. using redundant or spare channels or apparatus
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/12—Arrangements for detecting or preventing errors in the information received by using return channel
    • H04L1/16—Arrangements for detecting or preventing errors in the information received by using return channel in which the return channel carries supervisory signals, e.g. repetition request signals
    • H04L1/18—Automatic repetition systems, e.g. Van Duuren systems
    • H04L1/1809—Selective-repeat protocols
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/12—Arrangements for detecting or preventing errors in the information received by using return channel
    • H04L1/16—Arrangements for detecting or preventing errors in the information received by using return channel in which the return channel carries supervisory signals, e.g. repetition request signals
    • H04L1/18—Automatic repetition systems, e.g. Van Duuren systems
    • H04L1/1829—Arrangements specially adapted for the receiver end
    • H04L1/1835—Buffer management
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/12—Arrangements for detecting or preventing errors in the information received by using return channel
    • H04L1/16—Arrangements for detecting or preventing errors in the information received by using return channel in which the return channel carries supervisory signals, e.g. repetition request signals
    • H04L1/18—Automatic repetition systems, e.g. Van Duuren systems
    • H04L1/1867—Arrangements specially adapted for the transmitter end
    • H04L1/1874—Buffer management
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
    • H04L69/12—Protocol engines
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
    • H04L69/16—Implementation or adaptation of Internet protocol [IP], of transmission control protocol [TCP] or of user datagram protocol [UDP]
    • H04L69/161—Implementation details of TCP/IP or UDP/IP stack architecture; Specification of modified or new header fields
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
    • H04L69/16—Implementation or adaptation of Internet protocol [IP], of transmission control protocol [TCP] or of user datagram protocol [UDP]
    • H04L69/163—In-band adaptation of TCP data exchange; In-band control procedures
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
    • H04L69/40—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection

Definitions

  • the present invention is related generally to intelligent network adapters. More specifically, the present invention is directed toward the state recovery and failover of a network offload engine.
  • IP networks In an Internet Protocol (IP) Network, the software provides a message passing mechanism that can be used to communicate with Input/Output devices, general purpose computers (host), and special purpose computers.
  • IP networks generally employ what is known as a "protocol stack" to encapsulate and transmit data using a series of protocols.
  • a typical protocol stack following the Open System Interconnection (OSI) architectural model includes a link layer protocol, a network layer protocol, a transport protocol, an upper level protocol, and an application programming interface.
  • the key standard transport protocols used on IP networks today are the Transmission Control Protocol (TCP) and the User Datagram Protocol (UDP).
  • TCP Transmission Control Protocol
  • UDP User Datagram Protocol
  • SCTP Stream Control Transmission Protocol
  • Processes executing on devices or computers access the IP network through Upper Level Protocols, such as Sockets, iSCSI, and Direct Access File System (DAFS).
  • DAFS Direct Access File System
  • the host CPU generally performs most of the transport (e.g., TCP) and network layer (e.g., IP) processing.
  • the network interface card NIC
  • the link layer e.g., Ethernet
  • the host software maintains all the state information associated with TCP/IP connections in host local memory. Keeping all the state information in host local memory allows the host software to support switchover, and switchback, between a primary NIC and an alternate NIC. That is, if the primary NIC fails, the host software moves all the connections to the alternate NIC and continues communication processing.
  • Network offload engines offer a higher performance interface for communicating to other general purpose computers and I/O devices.
  • Network offload engines offload the transport (e.g., TCP) and network (e.g., IP) layer into the network offload engine.
  • transport e.g., TCP
  • network e.g., IP
  • US 6,049,825 discloses a method and system for switching between duplicated network adapters in order to provide a host computer with fault-tolerant functionality for TCP/IP communication with a plurality of other hosts over a network.
  • a first host is equipped with a first and second network adapters, which constructs a hot standby system.
  • the first host disconnects the first network adapter and activates the second network adapter.
  • the first host then initializes its own network table (or ARP cache) and initiates transmission of a control message addressed to the second host.
  • WO 03/021436 A2 discloses method and apparatus for the selective offloading of protocol processing.
  • computationally intensive and memory bandwidth intensive protocol processing tasks are offloaded from the host processor of a computer (12) to an auxiliary processor (152).
  • the present invention provides a data processing system for failover of a network offload engine in communication with a network and associated with a host, the method comprising: means for reconstructing state information for the network offload engine from information stored at a location external to the network offload engine; means for resuming network operation with a standby network adapter in accordance with the state information.
  • the system further comprises means for designating at least one selected network connection in the network offload engine as having failover support, wherein the state information is specific to the at least one selected network connection and network operation is resumed only with respect to the at least one selected network connection.
  • the at least one selected network connection is designated as having failover support by setting a socket option associated with the at least one selected network connection.
  • the location external to the network offload engine is the host. Still more preferably, the location external to the network offload engine is one of the standby network adapter and another network adapter.
  • information stored on the host is sufficient to reconstruct hard state information for the network offload engine.
  • the system further comprises: means for acquiring soft state information during the resumed network operation with the standby network adapter.
  • the hard state information includes at least one of a last sequence number received, a last sequence number sent, a last acknowledgement number received, and a last acknowledgement number sent.
  • the network offload engine when operable, sends an acknowledgment for received data only after the received data has been delivered to the host.
  • a transmit buffer in the host is released only after an acknowledgment for transmitted data associated with the transmit buffer has been received by the network offload engine, when the network offload engine is operable.
  • the network offload engine when operable, sends an acknowledgment for received data only after the received data has been delivered to the host; and wherein a transmit buffer in the host is released only after an acknowledgment for transmitted data associated with the transmit buffer has been received by the network offload engine, when the network offload engine is operable.
  • the system further comprises means for checkpointing, on a pre-defined periodic basis, at least a subset of state information associated with the network offload engine, when operable, by copying the at least a subset of state information associated with the network offload engine to memory in the host.
  • the system further comprises means for detecting failure of the network offload engine; means, responsive to detecting the failure of the network offload engine, for identifying connections associated with the network offload engine; designating the connections as recovering connections; recovering state information for each of the recovering connections; and resuming network operation of each of the recovering connections on the standby network adapter as the state information for each of the recovering connections is recovered.
  • the standby network adapter is a standby network offload engine.
  • the standby network adapter requires the host to perform protocol processing.
  • a data processing system comprising: a host; a network offload engine in communication with a network and associated with the host, the data processing system comprising; a standby network adapter in communication with the network and associated with the host; and a host control program associated with the host, wherein the host executes the host control program to perform acts including: reconstructing state information for the network offload engine from information stored in the host; resuming network operation with the standby network adapter in accordance with the state information.
  • the present invention provides a computer-based method for failover of a network offload engine in communication with a network and associated with a host, the method comprising: reconstructing state information for the network offload engine from information stored at a location external to the network offload engine; resuming network operation with a standby network adapter in accordance with the state information.
  • the present invention provides a computer program comprising program code means adapted to perform the steps of the method described above when said program is run on a computer.
  • the present invention provides a method, computer program product, and data processing system for efficiently recovering state and performing failover of a network offload engine.
  • the present invention distinguishes between the hard state and the soft state of a protocol.
  • Hard state is state information that, when lost, leads to incorrect protocol behavior.
  • Soft state is state information that may be lost or become inconsistent without loss of correctness.
  • the present invention ensures correctness by always being able to recover the hard state of the protocol.
  • a preferred embodiment of the present invention maintains allows the hard state of a reliable connection protocol, such as TCP, to be recovered by enforcing two invariants (i.e., conditions that always hold true) with regard to the normal course of communication between the host and adapter.
  • the net effect of these invariants is to allow the hard state to be recovered with less frequent checkpointing or synchronization of the host with the adapter, since hard state can be recovered from information normally maintained by the host.
  • a preferred embodiment of the present invention performs a failover of a network offload engine by temporarily blocking the reception of network packets, recovering hard state from host information, resuming network operation using a substitute network offload engine, and recovering soft state from the subsequent network activity.
  • Network data processing system 100 is a network of computers in which the present invention may be implemented.
  • Network data processing system 100 contains a network 102, which is the medium used to provide communications links between various devices and computers connected together within network data processing system 100.
  • Network 102 may include connections, such as wire, wireless communication links, or fiber optic cables.
  • server 104 is connected to network 102 along with storage unit 106.
  • clients 108, 110, and 112 are connected to network 102. These clients 108, 110, and 112 may be, for example, personal computers or network computers.
  • server 104 provides data, such as boot files, operating system images, and applications to clients 108-112.
  • Clients 108, 110, and 112 are clients to server 104.
  • Network data processing system 100 may include additional servers, clients, and other devices not shown.
  • network data processing system 100 is the Internet with network 102 representing a worldwide collection of networks and gateways that use the Transmission Control Protocol/Internet Protocol (TCP/IP) suite of protocols to communicate with one another.
  • TCP/IP Transmission Control Protocol/Internet Protocol
  • network data processing system 100 also may be implemented as a number of different types of networks, such as for example, an intranet, a local area network (LAN), or a wide area network (WAN).
  • Figure 1 is intended as an example, and not as an architectural limitation for the present invention.
  • Data processing system 200 may be a symmetric multiprocessor (SMP) system including a plurality of processors 202 and 204 connected to system bus 206. Alternatively, a single processor system may be employed. Also connected to system bus 206 is memory controller/cache 208, which provides an interface to local memory 209. I/O bus bridge 210 is connected to system bus 206 and provides an interface to I/O bus 212. Memory controller/cache 208 and I/O bus bridge 210 may be integrated as depicted.
  • SMP symmetric multiprocessor
  • Peripheral component interconnect (PCI) bus bridge 214 connected to I/O bus 212 provides an interface to PCI local bus 216.
  • PCI Peripheral component interconnect
  • a number of modems may be connected to PCI local bus 216.
  • Typical PCI bus implementations will support four PCI expansion slots or add-in connectors.
  • Communications links to clients 108-112 in Figure 1 may be provided through modem 218 and network adapter 220 connected to PCI local bus 216 through add-in boards.
  • network adapter 220 is a network offload engine, such as an RNIC (Remote direct memory access Network Interface Card).
  • RNIC Remote direct memory access Network Interface Card
  • Additional PCI bus bridges 222 and 224 provide interfaces for additional PCI local buses 226 and 228, from which additional modems or network adapters may be supported. In this manner, data processing system 200 allows connections to multiple network computers.
  • a memory-mapped graphics adapter 230 and hard disk 232 may also be connected to I/O bus 212 as depicted, either directly or indirectly.
  • the data processing system depicted in Figure 2 may be, for example, an IBM eServer pSeries system, a product of International Business Machines Corporation in Armonk, New York, running the Advanced Interactive Executive (AIX) operating system or LINUX operating system (IBM and AIX are registered trademarks of IBM Corporation; Linux is a trademark of Linus Torvalds in the United States, other countries, or both).
  • IBM eServer pSeries system a product of International Business Machines Corporation in Armonk, New York, running the Advanced Interactive Executive (AIX) operating system or LINUX operating system
  • AIX Advanced Interactive Executive
  • LINUX LINUX operating system
  • Data processing system 300 is an example of a client computer.
  • Data processing system 300 employs a peripheral component interconnect (PCI) local bus architecture.
  • PCI peripheral component interconnect
  • AGP Accelerated Graphics Port
  • ISA Industry Standard Architecture
  • Processor 302 and main memory 304 are connected to PCI local bus 306 through PCI bridge 308.
  • PCI bridge 308 also may include an integrated memory controller and cache memory for processor 302. Additional connections to PCI local bus 306 may be made through direct component interconnection or through add-in boards.
  • local area network (LAN) adapter 310 SCSI host bus adapter 312, and expansion bus interface 314 are connected to PCI local bus 306 by direct component connection.
  • LAN adapter 310 is a network offload engine, such as an RNIC (Remote direct memory access Network Interface Card).
  • audio adapter 316, graphics adapter 318, and audio/video adapter 319 are connected to PCI local bus 306 by add-in boards inserted into expansion slots.
  • Expansion bus interface 314 provides a connection for a keyboard and mouse adapter 320, modem 322, and additional memory 324.
  • Small computer system interface (SCSI) host bus adapter 312 provides a connection for hard disk drive 326, tape drive 328, and CD-ROM drive 330. Typical PCI local bus implementations will support three or four PCI expansion slots or add-in connectors.
  • An operating system runs on processor 302 and is used to coordinate and provide control of various components within data processing system 300 in Figure 3 .
  • the operating system may be a commercially available operating system, such as Windows XP, which is available from Microsoft Corporation (Microsoft, Windows, Windows NT, and the Windows logo are trademarks of Microsoft Corporation in the United States, other countries, or both).
  • An object oriented programming system such as Java may run in conjunction with the operating system and provide calls to the operating system from Java programs or applications executing on data processing system 300.
  • Java and all Java-based trademarks are trademarks of Sun Microsystems, Inc. in the United States, other countries, or both.
  • Instructions for the operating system, the object-oriented operating system, and applications or programs are located on storage devices, such as hard disk drive 326, and may be loaded into main memory 304 for execution by processor 302.
  • FIG. 3 may vary depending on the implementation.
  • Other internal hardware or peripheral devices such as flash read-only memory (ROM), equivalent nonvolatile memory, or optical disk drives and the like, may be used in addition to or in place of the hardware depicted in Figure 3 .
  • the processes of the present invention may be applied to a multiprocessor data processing system.
  • data processing system 300 may be a stand-alone system configured to be bootable without relying on some type of network communication interfaces
  • data processing system 300 may be a personal digital assistant (PDA) device, which is configured with ROM and/or flash ROM in order to provide non-volatile memory for storing operating system files and/or user-generated data.
  • PDA personal digital assistant
  • data processing system 300 also may be a notebook computer or hand held computer in addition to taking the form of a PDA.
  • data processing system 300 also may be a kiosk or a Web appliance.
  • the present invention provides a method, computer program product, and data processing system for efficiently recovering state and performing failover of a network offload engine.
  • TCP Transmission Control Protocol
  • FIG 4 is a diagram depicting the format of a TCP packet 400.
  • Figure 4 is adapted from W. Richard Stevens, TCP/IP Illustrated, Vol. 1, Addison-Wesley, 1994, p. 225 , Figure 17.2.
  • TCP packet 400 includes a 32-bit sequence number 402, a 32-bit acknowledgement number 404, and a number of flags 406.
  • sequence number 402 allows the receiving computer to reassemble the correct data in the correct order.
  • the receiving computer sends back a packet with acknowledgement number 404 containing the sequence number of the received packet and with the "ACK" flag from flags 406 set to denote an acknowledgement. This is called “sending an ACK.” If the sending computer does not receive an "ACK” (e.g., because the packet was lost or garbled), the sending computer will resend the packet until it receives an ACK. If the packets are received out of order (e.g., a packet with sequence number 5 precedes a packet with sequence number 4), the receiving computer simply waits until a continuous series of packets can be assembled before making the received data available to applications.
  • TCP and other similar reliable connection protocols are state-aware. That is, TCP relies on the keeping of state information, such as the last sequence number received or the last sequence number sent. In general, if the state information is lost, TCP can no longer function. Thus, if a network offload engine/adapter providing a state-aware protocol, such as TCP, fails in the middle of a network connection, if the state information contained in the adapter is completely lost, the connection is permanently lost.
  • State information may be preserved by way of "checkpointing.”
  • the state information contained in the adapter may periodically be relayed to the adapter's host computer. In high speed operation, however, this relaying process may result in a performance bottleneck, since some protocol events require immediate checkpointing in order to preserve correctness (i.e., in order for state information to be correctly recovered in the event of a failure).
  • the present invention is directed toward reducing this performance bottleneck.
  • the present invention addresses this problem in three important ways.
  • First, the present invention distinguishes between "hard” and “soft” state information (q.v.), recognizing that for failover, only "hard” state information need be maintained.
  • Second, the present invention provides a method of deriving hard state information from the circumstances surrounding the data transfer operations that are observable to the host.
  • Third, the present invention because it allows checkpointing to take place on a discretionary basis, allows a designer to tune an adapter/host design to meet desired performance requirements by varying the frequency at which checkpointing occurs.
  • FIG. 5 is a diagram providing an overall conceptual view of a preferred embodiment of the present invention.
  • Host 500 is associated with a TCP/IP offload engine (intelligent network adapter) 501.
  • Offload engine 501 communicates with host 500 primarily via direct memory access (DMA) 507, to allow offload engine 501 to send and receive data to/from host memory with consuming minimal processor time at host 500.
  • DMA direct memory access
  • Host control software 502 provides an interface from applications to offload engine 501. Host control software 502 can also initiate a failover operation in the event offload engine 501 fails.
  • per-connection information 504 is stored in host 500.
  • Per-connection information 504 includes state information 506 regarding the connection, which may include information regarding TCP sequence and acknowledgement numbers, connection parameters, and the like.
  • per-connection information 504 also includes a designation 505 of whether a given connection will be given failover support in the event offload engine 501 fails. This allows a choice to be made as to which connections will be preserved when and if a failover to another offload engine occurs.
  • designation 505 may be set through the use of a socket option.
  • a "setsockopt" function is defined to allow a socket option associated with a given connection to be set to a particular value.
  • Buffers 508 are also includes with per-connection information 504.
  • Data is transferred between host 500 and offload engine 501 via DMA 507.
  • host 500 allocates a buffer and copies the data to be transmitted into the allocated buffer.
  • Offload engine 501 copies the contents of the allocated buffer into one of offload engine 501's own buffers 510 in offload engine 501's own per-connection information 509.
  • the data is placed in one of offload engine 501 's buffers 510 and copied, via DMA 507, to one of buffers 508.
  • buffers 508 and 510 may include both transmit and receive buffers.
  • Offload engine 501 has its own per-connection information 509, including buffers 510 and TCP state information 512 .
  • Offload engine control software 514 directs the transfer of data from to/from host 500 via DMA 507 and the transfer of data via TCP/IP protocol 516 over network 518.
  • offload engine In general, the terms "offload engine,” “adapter,” and “card” are herein used interchangeably unless specifically noted. An important exception to the general rule is in the case of a standby network adapter. While a preferred embodiment of the present invention uses a network offload engine as the standby adapter, one of ordinary skill in the art will recognize that a conventional "dumb” network adapter may be substituted for a network offload engine, while allowing the host to perform any protocol processing (e.g., TCP/IP processing).
  • protocol processing e.g., TCP/IP processing
  • hard state is defined as that state which may not be lost without losing correctness
  • soft state is state information that is non-essential for maintaining correctness inasmuch as losing it only translates to a performance penalty.
  • Hard state typically consists of items that are required for reliable data transfer such as sequence numbers and data buffers, and those required for connection management, such as IP addresses and negotiated options.
  • Soft state is usually information that is learned over the lifetime of the connection, such as the round-trip time, the congestion window, etc. The implication is that, when an adapter fails, as long as no hard state is lost, recovering from that failure will be possible. Performance may be affected, in that soft state will need to be regenerated, but correctness will be maintained.
  • Both transmit and receive buffers are hard state. Receive buffers are hard state since once they have been acknowledged to the other end, the other end cannot be expected to maintain a copy, and thus will not be able to retransmit it. Transmit buffers are hard state because once an application gives data to a reliable protocol, the host is not expected to maintain a copy; the application expects all retransmissions to occur underneath transparently. Hence for failover, such buffers must be retained in host memory. (Note that it is also possible to retain these buffers in the memory of another adapter, instead of host memory.)
  • Some hard state values such as the network address of the remote host or negotiated options, are as a matter of course stored by the host, so recovery of these values is necessary.
  • Sequence and ACK numbers are hard state in the TCP stack, but do need to be recovered.
  • the four key state variables in this case are "last sequence number sent”, “last sequence number received”, “last ACK number sent” and “last ACK number received”. Keeping these numbers synchronized between the host and the adapter every time they are updated could be prohibitively expensive and might defeat the performance gains of offload.
  • Condition 2A ensures that, if the adapter fails, the host will always have those buffers to retransmit for which it has not seen ACKs yet.
  • Condition 2B ensures that if the card failed without transferring some received data to the host, the remote end will always be able to retransmit that data, since the remote end is not "ACKed" until the host sees the data.
  • the adapter can synchronize the remaining variables with the host "lazily" at some chosen interval. This allows the adapter to batch several updates into one update, and avoid the performance overhead of frequent synchronization. How frequently the adapter returns new ACK values to the host now only becomes a performance issue and not a correctness issue. Note that there is a performance-space tradeoff in choosing the synchronization period: the longer this interval is, the fewer I/O transfers occur between the card and the host, but the host must hold on to buffers longer. A shorter interval, on the other hand, means the host can free buffers as soon as possible, but greater I/O overhead will be incurred. For performance reasons, the first case is clearly preferable, and thus the hope is that the host has enough memory to allow the card to "lazily" report acknowledgments.
  • Figures 6A-6B and 7A-7B contrast normal TCP operation on an established connection as known in the art with the application of conditions 2A and 2B to an established TCP connection as may be achieved in practice in accordance with a preferred embodiment of the present invention.
  • An established TCP connection is said to be in the "ESTABLISHED" state (see ESTABLISHED state 800 in the TCP state diagram provided in Figure 8 ).
  • Figure 6A is a diagram depicting a conventional approach to TCP transmission.
  • Adapter 601 copies data to be sent from buffer 602 on host 600 into adapter 601's own buffer 604. At this point host 600 will deallocate 602, as it is no longer needed.
  • Adapter 601 processes (TCP processing 616) the data that is now in buffer 604 to place it in TCP packet form for transmission 608. Transmission 608 is repeated until an ACK corresponding to the sent data is received by adapter 601.
  • FIG. 6B is a diagram depicting TCP transmission in accordance with a preferred embodiment of the present invention.
  • Adapter 611 copies data from buffer 612 on host 610 into adapter 611's buffer 614, but buffer 612 is not deallocated.
  • TCP processing 616 is performed and the data is transmitted over the network (transmission 618).
  • an ACK is received (ACK reception 620)
  • an ACK notification message 622 is generated in adapter 611.
  • adapter 611 transmits ACK notification 622 to host 610, where host 610 processes (ACK notification processing 624 ) ACK notification 622 and deallocates buffer 612 in response.
  • the ACK notifications may be stored in adapter 611 temporarily and then processed in a batch fashion for greater efficiency.
  • FIG. 7A is a diagram depicting a conventional approach to TCP reception.
  • Adapter 700 receives a TCP data packet (data reception 702) from the network.
  • the packet is processed (TCP packet processing 704) to obtain the data stored in the packet, which is copied into buffer 706 on adapter 700 .
  • TCP packet processing 704 also results in the generation of a corresponding ACK to the packet.
  • This ACK packet is transmitted back to the sender of the original received packet (ACK transmission 708 ).
  • buffer 706 is then copied (e.g., via DMA) into a buffer 710 on host 712 and buffer 706 is deallocated. This copying of buffers from adapter 700 to host 712 may take place in a batch process.
  • FIG. 7B is a diagram depicting TCP reception in accordance with a preferred embodiment of the present invention.
  • Adapter 720 receives a TCP data packet from the network (data reception 722).
  • the packet is processed (TCP packet processing 724) to obtain the data stored in the packet, which is copied into buffer 726 on adapter 720, but no ACK is generated yet.
  • buffer 726 is then copied (e.g., via DMA) into a buffer 728 on host 730 and buffer 726 is deallocated. Again, this copying of buffers from adapter 720 to host 730 may take place in a batch process.
  • a corresponding ACK is then generated and transmitted to original sender of the packet (ACK transmission 732).
  • FIG. 8 The full TCP state diagram depicting ESTABLISHED state 800 and the other TCP states is provided in Figure 8.
  • Figure 8 is adapted from W. Richard Stevens, TCP/IP Illustrated, Vol. 1, Addison-Wesley, 1994, p. 241 , Figure 18.12.
  • the names for the TCP states are those described in RFC (Request for Comments) 793, in which the TCP protocol is described.
  • RFC Request for Comments
  • the host with the offload engine/adapter will be referred to as the "offload host” and the remote end as the "remote host”. It should be noted, for conceptual clarity, that because the host and adapter both keep track of state information, the host has a state that it believes the connection is in and the adapter also has a state that it believes the connection is in. The goal of the present invention is to ensure that even if the host and adapter are not perfectly synchronized, the host is in a state from which recovery is possible.
  • FIGS 9A-9D are a series of diagrams depicting a failover process executed in accordance with a preferred embodiment of.the present invention.
  • adapter failure must be detected.
  • stage 1 (“Detect card failure") in Figure 9A
  • host 900 which maintains a number of connections (denoted by per-connection state information blocks 902) utilizing an adapter 904.
  • Host 900 first detects that adapter 904 has failed. This can be done in a number of ways that are known in the art.
  • One particular method that may be employed is for host 900 and adapter 904 to send period "heartbeat" messages to each other. If host 900 times out waiting for a heartbeat message to arrive from adapter 904, host 900 can safely assume that adapter 904 has failed and that a standby adapter 906 should replace failed adapter 904.
  • stage 2 (“Blocking phase") in Figure 9B , to begin the failover/recovery process, host 900 engages standby adapter 906, but blocks all packets from the network (blocking operation 908 ).
  • Host 900 marks the state of each connection handled by failed adapter 904 (per-connection information 902) to indicate that these connections are in the "RECOVERING" state.
  • stage 3 (“State reconstruction phase") in Figure 9C , host 900 now unblocks the packets from the network and begins the process of recovering connections. In this stage, packets arriving to any connection in the RECOVERING state are dropped. All other connections are processed according to normal TCP processing. The following sequence of operations describes the per-connection recovery process:
  • connection offload engine Once a connection has been recovered, normal operation for that connection can resume at standby adapter 906, as shown in stage 4 ("Resume normal operation") in Figure 9D .
  • host 900 could take over TCP/IP processing, rather than using an intelligent standby adapter (network offload engine).
  • Figure 10 is a flowchart representation of a process of performing a failover of a network offload engine in accordance with a preferred embodiment of the present invention.
  • the host detects an adapter failure (block 1000 ).
  • the host recreates whatever hard state information is not already stored in the host (block 1002 ).
  • This hard state information is propagated to a standby network offload engine (or retained in the host, if a "dumb" standby adapter is used) (block 1004 ).
  • Network operation is resumed using the standby adapter (block 1006 ).
  • any soft state information can be reacquired through observation of the network connection in practice (block 1008 ).
  • the computer readable media may take the form of coded formats that are decoded for actual use in a particular data processing system.
  • Functional descriptive material is information that imparts functionality to a machine.
  • Functional descriptive material includes, but is not limited to, computer programs, instructions, rules, facts, definitions of computable functions, objects, and data structures.

Claims (12)

  1. Datenverarbeitungssystem zum ausfallbedingten Wechsel (failover) von einer Netzwerk-Entlastungsmaschine (501), wobei die Netzwerk-Entlastungsmaschine mit einem Netzwerk Daten austauscht und zu einem Hostrechner gehört, zu einem Bereitschaftsmodus-Adapter (601) nach dem Ausfall der Netzwerk-Entlastungsmaschine, wobei das System Folgendes umfasst:
    ein Mittel, um Zustandsinformationen eines Zustandserkennungsprotokolls für die Netzwerk-Entlastungsmaschine aus Informationen wiederherzustellen, die an einem Ort gespeichert werden, der sich außerhalb der Netzwerk-Entlastungsmaschine befindet; und
    ein Mittel, um den Netzwerkbetrieb mit Hilfe des Bereitschaftsmodus-Netzwerkadapters entsprechend den wiederhergestellten Zustandsinformationen, die zu der Netzwerk-Entlastungsmaschine gehören, wieder aufzunehmen.
  2. Datenverarbeitungssystem nach Anspruch 1, das des Weiteren Folgendes umfasst:
    ein Mittel, um mindestens eine ausgewählte Netzwerkverbindung in der Netzwerk-Entlastungsmaschine als eine Netzwerkverbindung zu kennzeichnen, die über eine Ausfallsicherungs-Unterstützung verfügt,
    wobei die Zustandsinformationen für die mindestens eine ausgewählte Netzwerkverbindung spezifisch sind und der Netzwerkbetrieb nur in Bezug auf die mindestens eine ausgewählte Netzwerkverbindung wieder aufgenommen wird.
  3. Datenverarbeitungssystem nach Anspruch 1 oder Anspruch 2, wobei auf dem Hostrechner gespeicherte Informationen ausreichend sind, um Informationen über harte Zustände für die Netzwerk-Entlastungsmaschine wiederherzustellen.
  4. Datenverarbeitungssystem nach Anspruch 3, das des Weiteren Folgendes umfasst:
    ein Mittel, um Informationen über weiche Zustände während des wieder aufgenommenen Netzwerkbetriebs mit dem Bereitschaftsmodus-Netzwerkadapter zu erfassen.
  5. Datenverarbeitungssystem nach Anspruch 3 oder Anspruch 4, wobei die Informationen über harte Zustände mindestens eine der folgenden Nummern enthalten: eine empfangene letzte Folgenummer, eine gesendete letzten Folgenummer, eine empfangene letzte Bestätigungsnummer oder eine gesendete letzte Bestätigungsnummer
  6. Datenverarbeitungssystem nach einem der Ansprüche 3 bis 5, wobei die Netzwerk-Entlastungsmaschine, wenn sie betriebsbereit ist, eine Bestätigung für empfangene Daten erst sendet, nachdem die empfangenen Daten an den Hostrechner geliefert worden sind.
  7. Datenverarbeitungssystem nach einem der Ansprüche 3 bis 6, wobei ein Sendepufferspeicher in dem Hostrechner erst freigegeben wird, nachdem eine Bestätigung für übertragene Daten, die zu dem Sendepufferspeicher gehören, von der Netzwerk-Entlastungsmaschine empfangen worden ist, wenn die Netzwerk-Entlastungsmaschine betriebsbereit ist.
  8. Datenverarbeitungssystem nach einem der Ansprüche 3 bis 7, wobei die Netzwerk-Entlastungsmaschine, wenn sie betriebsbereit ist, eine Bestätigung für empfangene Daten erst sendet, nachdem die empfangenen Daten an den Hostrechner geliefert worden sind; und wobei ein Sendepufferspeicher in dem Hostrechner erst freigegeben wird, nachdem eine Bestätigung für übertragene Daten, die zu dem Sendepufferspeicher gehören, von der Netzwerk-Entlastungsmaschine empfangen worden ist, wenn die Netzwerk-Entlastungsmaschine betriebsbereit ist.
  9. Datenverarbeitungssystem nach einem der vorhergehenden Ansprüche, das des Weiteren Folgendes umfasst:
    ein Mittel, um auf einer vorher festgelegten, in regelmäßigen Abständen wiederkehrenden Basis mindestens eine Teilmenge an Zustandsinformationen, die zu der Netzwerk-Entlastungsmaschine, wenn sie betriebsbereit ist, gehören, mit einem Prüfpunkt zu versehen, indem die mindestens eine Teilmenge an Zustandsinformationen, die zu der Netzwerk-Entlastungsmaschine gehören, in den Speicher in dem Hostrechner kopiert werden.
  10. Datenverarbeitungssystem nach einem der vorhergehenden Ansprüche, das des Weiteren Folgendes umfasst:
    ein Mittel, um einen Ausfall der Netzwerk-Entlastungsmaschine festzustellen (1000);
    ein Mittel, das auf das Feststellen des Ausfalls der Netzwerk-Entlastungsmaschine reagiert, um Verbindungen zu ermitteln, die zu der Netzwerk-Entlastungsmaschine gehören;
    Kennzeichnen der Verbindungen als Wiederherstellungs-Verbindungen;
    Wiederherstellen (1002) von Zustandsinformationen für jede der Wiederherstellungs-Verbindungen; und
    Wiederaufnehmen (1006) des Netzwerkbetriebs einer jeden der Wiederherstellungs-Verbindungen auf dem Bereitschaftsmodus-Netzwerkadapter, wenn die Zustandsinformationen für jede der Wiederherstellungs-Verbindungen wiederhergestellt werden.
  11. Verfahren zum ausfallbedingten Wechsel von einer Netzwerk-Entlastungsmaschine, wobei die Netzwerk-Entlastungsmaschine mit einem Netzwerk Daten austauscht und zu einem Hostrechner gehört, zu einem Bereitschaftsmodus-Adapter nach dem Ausfall der Netzwerk-Entlastungsmaschine, wobei das Verfahren Folgendes umfasst:
    Wiederherstellen von Zustandsinformationen eines Zustandserkennungsprotokolls für die Netzwerk-Entlastungsmaschine aus Informationen, die an einem Ort gespeichert werden, der sich außerhalb der Netzwerk-Entlastungsmaschine befindet; und
    Wiederaufnehmen des Netzwerkbetriebs mit dem Bereitschaftsmodus-Netzwerkadapter entsprechend den wiederhergestellten Zustandsinformationen, die zu der Netzwerk-Entlastungsmaschine gehören.
  12. Rechnerprogramm, das ein Programmcode-Mittel umfasst, das so ausgelegt ist, dass es alle Schritte nach Anspruch 11 durchführt, wenn das Programm auf einem Rechner ausgeführt wird.
EP03816813A 2003-04-02 2003-07-21 Zustandswiederherstellung und failover intelligenter netzwerkadapter Expired - Lifetime EP1609063B1 (de)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US406095 1982-08-06
US10/406,095 US7114096B2 (en) 2003-04-02 2003-04-02 State recovery and failover of intelligent network adapters
PCT/US2003/022789 WO2004097641A1 (en) 2003-04-02 2003-07-21 State recovery and failover of intelligent network adapters

Publications (3)

Publication Number Publication Date
EP1609063A1 EP1609063A1 (de) 2005-12-28
EP1609063A4 EP1609063A4 (de) 2009-03-11
EP1609063B1 true EP1609063B1 (de) 2011-08-31

Family

ID=33097247

Family Applications (1)

Application Number Title Priority Date Filing Date
EP03816813A Expired - Lifetime EP1609063B1 (de) 2003-04-02 2003-07-21 Zustandswiederherstellung und failover intelligenter netzwerkadapter

Country Status (6)

Country Link
US (1) US7114096B2 (de)
EP (1) EP1609063B1 (de)
CN (1) CN100399282C (de)
AT (1) ATE522865T1 (de)
AU (1) AU2003259192A1 (de)
WO (1) WO2004097641A1 (de)

Families Citing this family (85)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6850987B1 (en) * 1999-06-01 2005-02-01 Fastforward Networks, Inc. System for multipoint infrastructure transport in a computer network
US7447795B2 (en) * 2001-04-11 2008-11-04 Chelsio Communications, Inc. Multi-purpose switching network interface controller
US8312117B1 (en) * 2001-11-15 2012-11-13 Unisys Corporation Dialog recovery in a distributed computer system
US7644436B2 (en) * 2002-01-24 2010-01-05 Arxceo Corporation Intelligent firewall
US7007103B2 (en) * 2002-04-30 2006-02-28 Microsoft Corporation Method to offload a network stack
US7275081B1 (en) 2002-06-10 2007-09-25 Juniper Networks, Inc. Managing state information in a computing environment
US20050022017A1 (en) * 2003-06-24 2005-01-27 Maufer Thomas A. Data structures and state tracking for network protocol processing
US7739403B1 (en) 2003-10-03 2010-06-15 Juniper Networks, Inc. Synchronizing state information between control units
US7668923B2 (en) * 2003-10-14 2010-02-23 International Business Machines Corporation Master-slave adapter
US7864806B2 (en) * 2004-01-06 2011-01-04 Broadcom Corp. Method and system for transmission control packet (TCP) segmentation offload
US7668165B2 (en) * 2004-03-31 2010-02-23 Intel Corporation Hardware-based multi-threading for packet processing
JP4343760B2 (ja) * 2004-04-28 2009-10-14 株式会社日立製作所 ネットワークプロトコル処理装置
US7831745B1 (en) * 2004-05-25 2010-11-09 Chelsio Communications, Inc. Scalable direct memory access using validation of host and scatter gather engine (SGE) generation indications
US7761529B2 (en) * 2004-06-30 2010-07-20 Intel Corporation Method, system, and program for managing memory requests by devices
US20060165084A1 (en) * 2005-01-21 2006-07-27 International Business Machines Corporation RNIC-BASED OFFLOAD OF iSCSI DATA MOVEMENT FUNCTION BY TARGET
US20060259570A1 (en) * 2005-05-13 2006-11-16 Microsoft Corporation Method and system for closing an RDMA connection
US7715436B1 (en) 2005-11-18 2010-05-11 Chelsio Communications, Inc. Method for UDP transmit protocol offload processing with traffic management
US7660264B1 (en) 2005-12-19 2010-02-09 Chelsio Communications, Inc. Method for traffic schedulign in intelligent network interface circuitry
US7724658B1 (en) 2005-08-31 2010-05-25 Chelsio Communications, Inc. Protocol offload transmit traffic management
US7660306B1 (en) 2006-01-12 2010-02-09 Chelsio Communications, Inc. Virtualizing the operation of intelligent network interface circuitry
US7616563B1 (en) 2005-08-31 2009-11-10 Chelsio Communications, Inc. Method to implement an L4-L7 switch using split connections and an offloading NIC
US7747999B1 (en) 2005-09-26 2010-06-29 Juniper Networks, Inc. Software installation in a multi-chassis network device
US7760733B1 (en) 2005-10-13 2010-07-20 Chelsio Communications, Inc. Filtering ingress packets in network interface circuitry
US7518986B1 (en) 2005-11-16 2009-04-14 Juniper Networks, Inc. Push-based hierarchical state propagation within a multi-chassis network device
US20070162639A1 (en) * 2005-11-30 2007-07-12 Chu Hsiao-Keng J TCP-offload-engine based zero-copy sockets
US7804769B1 (en) * 2005-12-01 2010-09-28 Juniper Networks, Inc. Non-stop forwarding in a multi-chassis router
US20070174723A1 (en) * 2006-01-18 2007-07-26 Omar Cardona Sub-second, zero-packet loss adapter failover
US20070165520A1 (en) * 2006-01-18 2007-07-19 Messing Jeffrey P Port trunking between switches
CN101022451B (zh) * 2006-02-14 2014-07-23 杭州华三通信技术有限公司 数据通信中连接状态的同步方法及其应用的通信节点
US20070233886A1 (en) * 2006-04-04 2007-10-04 Fan Kan F Method and system for a one bit TCP offload
US20070297334A1 (en) * 2006-06-21 2007-12-27 Fong Pong Method and system for network protocol offloading
US20080134300A1 (en) 2006-07-08 2008-06-05 David Izatt Method for Improving Security of Computer Networks
JP4776471B2 (ja) * 2006-08-11 2011-09-21 株式会社日立製作所 記憶システムおよび記憶システムの論理ユニット調査方法
US7571343B1 (en) * 2006-08-31 2009-08-04 Nortel Networks Limited Handling sequence numbers and/or an anti-replay window after failover between servers
US7821973B2 (en) * 2006-10-24 2010-10-26 Hewlett-Packard Development Company, L.P. Sharing of host bus adapter context
US7715321B2 (en) * 2007-01-30 2010-05-11 International Business Machines Corporation Network interface card transmission control protocol acceleration offload failure detection and recovery mechanism
US7929418B2 (en) * 2007-03-23 2011-04-19 Hewlett-Packard Development Company, L.P. Data packet communication protocol offload method and system
US7743160B2 (en) * 2007-03-29 2010-06-22 Blue Coat Systems, Inc. System and method of delaying connection acceptance to support connection request processing at layer-7
US8935406B1 (en) 2007-04-16 2015-01-13 Chelsio Communications, Inc. Network adaptor configured for connection establishment offload
US8589587B1 (en) 2007-05-11 2013-11-19 Chelsio Communications, Inc. Protocol offload in intelligent network adaptor, including application level signalling
US8060644B1 (en) 2007-05-11 2011-11-15 Chelsio Communications, Inc. Intelligent network adaptor with end-to-end flow control
US7826350B1 (en) 2007-05-11 2010-11-02 Chelsio Communications, Inc. Intelligent network adaptor with adaptive direct data placement scheme
US7831720B1 (en) 2007-05-17 2010-11-09 Chelsio Communications, Inc. Full offload of stateful connections, with partial connection offload
US20090064287A1 (en) 2007-08-28 2009-03-05 Rohati Systems, Inc. Application protection architecture with triangulated authorization
US7818606B1 (en) * 2007-09-28 2010-10-19 Emc Corporation Methods and apparatus for switch-initiated trespass decision making
US7962587B2 (en) * 2007-12-10 2011-06-14 Oracle America, Inc. Method and system for enforcing resource constraints for virtual machines across migration
US8086739B2 (en) * 2007-12-10 2011-12-27 Oracle America, Inc. Method and system for monitoring virtual wires
US7984123B2 (en) * 2007-12-10 2011-07-19 Oracle America, Inc. Method and system for reconfiguring a virtual network path
US8095661B2 (en) * 2007-12-10 2012-01-10 Oracle America, Inc. Method and system for scaling applications on a blade chassis
US7945647B2 (en) * 2007-12-10 2011-05-17 Oracle America, Inc. Method and system for creating a virtual network path
US8370530B2 (en) * 2007-12-10 2013-02-05 Oracle America, Inc. Method and system for controlling network traffic in a blade chassis
US7970951B2 (en) * 2008-02-29 2011-06-28 Oracle America, Inc. Method and system for media-based data transfer
US7965714B2 (en) * 2008-02-29 2011-06-21 Oracle America, Inc. Method and system for offloading network processing
US7944923B2 (en) * 2008-03-24 2011-05-17 Oracle America, Inc. Method and system for classifying network traffic
US20090288104A1 (en) * 2008-05-19 2009-11-19 Rohati Systems, Inc. Extensibility framework of a network element
US8677453B2 (en) * 2008-05-19 2014-03-18 Cisco Technology, Inc. Highly parallel evaluation of XACML policies
US8094560B2 (en) * 2008-05-19 2012-01-10 Cisco Technology, Inc. Multi-stage multi-core processing of network packets
US8667556B2 (en) * 2008-05-19 2014-03-04 Cisco Technology, Inc. Method and apparatus for building and managing policies
US7917614B2 (en) * 2008-06-10 2011-03-29 International Business Machines Corporation Fault tolerance in a client side pre-boot execution
US7941539B2 (en) * 2008-06-30 2011-05-10 Oracle America, Inc. Method and system for creating a virtual router in a blade chassis to maintain connectivity
US8739179B2 (en) * 2008-06-30 2014-05-27 Oracle America Inc. Method and system for low-overhead data transfer
US20100070471A1 (en) * 2008-09-17 2010-03-18 Rohati Systems, Inc. Transactional application events
JP5516569B2 (ja) * 2009-02-20 2014-06-11 富士通株式会社 情報処理装置及び制御方法
US8363549B1 (en) 2009-09-02 2013-01-29 Juniper Networks, Inc. Adaptively maintaining sequence numbers on high availability peers
US20110134930A1 (en) * 2009-12-09 2011-06-09 Mclaren Moray Packet-based networking system
US8713362B2 (en) 2010-12-01 2014-04-29 International Business Machines Corporation Obviation of recovery of data store consistency for application I/O errors
US8694821B2 (en) 2010-12-03 2014-04-08 International Business Machines Corporation Generation of standby images of applications
US8634415B2 (en) 2011-02-16 2014-01-21 Oracle International Corporation Method and system for routing network traffic for a blade server
US9858241B2 (en) 2013-11-05 2018-01-02 Oracle International Corporation System and method for supporting optimized buffer utilization for packet processing in a networking device
US8990560B2 (en) * 2011-06-17 2015-03-24 The Boeing Company Multiple independent levels of security (MILS) host to multilevel secure (MLS) offload communications unit
US9473596B2 (en) 2011-09-27 2016-10-18 International Business Machines Corporation Using transmission control protocol/internet protocol (TCP/IP) to setup high speed out of band data communication connections
US8995261B2 (en) * 2011-10-17 2015-03-31 Lg Electronics Inc. Method and apparatus of network traffic offloading
US8930507B2 (en) 2012-06-12 2015-01-06 International Business Machines Corporation Physical memory shared among logical partitions in a VLAN
US8880935B2 (en) * 2012-06-12 2014-11-04 International Business Machines Corporation Redundancy and load balancing in remote direct memory access communications
US9396101B2 (en) 2012-06-12 2016-07-19 International Business Machines Corporation Shared physical memory protocol
CN103944691B (zh) * 2013-01-17 2019-01-15 中兴通讯股份有限公司 一种协同业务传输中的数据重传方法及接入网网关
US9712541B1 (en) * 2013-08-19 2017-07-18 The Boeing Company Host-to-host communication in a multilevel secure network
US8984173B1 (en) 2013-09-26 2015-03-17 International Business Machines Corporation Fast path userspace RDMA resource error detection
US9489327B2 (en) 2013-11-05 2016-11-08 Oracle International Corporation System and method for supporting an efficient packet processing model in a network environment
WO2016019172A1 (en) * 2014-07-30 2016-02-04 Forward Networks, Inc. Systems and methods for network management
US10419170B2 (en) * 2015-02-26 2019-09-17 Qualcomm Incorporated RRC aware TCP retransmissions
US20160323792A1 (en) * 2015-04-29 2016-11-03 Aruba Networks, Inc. Wireless client session continuity across controller failover and load-balancing
US10649950B2 (en) * 2016-08-29 2020-05-12 Excelero Storage Ltd. Disk access operation recovery techniques
US11494125B2 (en) * 2020-12-17 2022-11-08 Western Digital Technologies, Inc. Storage system and method for dual fast release and slow release responses
US20220206908A1 (en) * 2020-12-30 2022-06-30 Oracle International Corporation Techniques for replicating state information for high availability

Family Cites Families (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5295258A (en) * 1989-12-22 1994-03-15 Tandem Computers Incorporated Fault-tolerant computer system with online recovery and reintegration of redundant components
US5774640A (en) * 1991-10-21 1998-06-30 Tandem Computers Incorporated Method and apparatus for providing a fault tolerant network interface controller
US5473599A (en) * 1994-04-22 1995-12-05 Cisco Systems, Incorporated Standby router protocol
KR100195065B1 (ko) * 1996-06-20 1999-06-15 유기범 데이타 통신망 정합장치
CA2216980C (en) * 1996-10-04 2001-09-25 Hitachi, Ltd. Communication method
US6049825A (en) * 1997-03-19 2000-04-11 Fujitsu Limited Method and system for switching between duplicated network interface adapters for host computer communications
US6134678A (en) * 1997-05-13 2000-10-17 3Com Corporation Method of detecting network errors
US6253334B1 (en) * 1997-05-13 2001-06-26 Micron Electronics, Inc. Three bus server architecture with a legacy PCI bus and mirrored I/O PCI buses
US6105151A (en) * 1997-05-13 2000-08-15 3Com Corporation System for detecting network errors
US6052733A (en) * 1997-05-13 2000-04-18 3Com Corporation Method of detecting errors in a network
US6314525B1 (en) * 1997-05-13 2001-11-06 3Com Corporation Means for allowing two or more network interface controller cards to appear as one card to an operating system
US6393483B1 (en) * 1997-06-30 2002-05-21 Adaptec, Inc. Method and apparatus for network interface card load balancing and port aggregation
US5937169A (en) * 1997-10-29 1999-08-10 3Com Corporation Offload of TCP segmentation to a smart adapter
US6157965A (en) * 1998-02-27 2000-12-05 Intel Corporation System and method for binding a virtual device driver to a network driver interface
US6311288B1 (en) * 1998-03-13 2001-10-30 Paradyne Corporation System and method for virtual circuit backup in a communication network
US6247141B1 (en) * 1998-09-24 2001-06-12 Telefonaktiebolaget Lm Ericsson (Publ) Protocol for providing replicated servers in a client-server system
US6618819B1 (en) * 1999-12-23 2003-09-09 Nortel Networks Limited Sparing system and method to accommodate equipment failures in critical systems
US6763479B1 (en) * 2000-06-02 2004-07-13 Sun Microsystems, Inc. High availability networking with alternate pathing failover
US6615221B2 (en) * 2001-03-09 2003-09-02 Hewlett-Packard Development Company, Lp. Scalable transport layer protocol for multiprocessor interconnection networks that tolerates interconnection component failure
US6853617B2 (en) * 2001-05-09 2005-02-08 Chiaro Networks, Ltd. System and method for TCP connection protection switching
US20030046330A1 (en) * 2001-09-04 2003-03-06 Hayes John W. Selective offloading of protocol processing
US6910150B2 (en) * 2001-10-15 2005-06-21 Dell Products L.P. System and method for state preservation in a stretch cluster
FR2831743B1 (fr) * 2001-10-25 2004-01-30 Cit Alcatel Systeme de routage is-is tolerant aux fautes et procede correspondant
US7535913B2 (en) * 2002-03-06 2009-05-19 Nvidia Corporation Gigabit ethernet adapter supporting the iSCSI and IPSEC protocols
US7292535B2 (en) * 2002-05-23 2007-11-06 Chiaro Networks Ltd Highly-available OSPF routing protocol
US7155632B2 (en) * 2002-06-27 2006-12-26 Nokia, Inc. Method and system for implementing IS-IS protocol redundancy
US7269133B2 (en) * 2002-09-03 2007-09-11 Jeremy Benjamin IS-IS high availability design
US6721806B2 (en) * 2002-09-05 2004-04-13 International Business Machines Corporation Remote direct memory access enabled network interface controller switchover and switchback support
US7599285B2 (en) * 2003-11-03 2009-10-06 Cisco Technology, Inc. Combined electro-mechanical and solid state switching fabric

Also Published As

Publication number Publication date
EP1609063A4 (de) 2009-03-11
EP1609063A1 (de) 2005-12-28
US7114096B2 (en) 2006-09-26
ATE522865T1 (de) 2011-09-15
AU2003259192A1 (en) 2004-11-23
US20040199808A1 (en) 2004-10-07
CN1739098A (zh) 2006-02-22
WO2004097641A1 (en) 2004-11-11
CN100399282C (zh) 2008-07-02

Similar Documents

Publication Publication Date Title
EP1609063B1 (de) Zustandswiederherstellung und failover intelligenter netzwerkadapter
US7213063B2 (en) Method, apparatus and system for maintaining connections between computers using connection-oriented protocols
Watson Timer-based mechanisms in reliable transport protocol connection management
Alvisi et al. Wrapping server-side TCP to mask connection failures
Peterson et al. Preserving and using context information in interprocess communication
US7469296B2 (en) Method and apparatus for an improved bulk read socket call
EP1665051B1 (de) Transparentes tcp-verbindungs-failover
US7107481B2 (en) Server takeover system and method
US7089289B1 (en) Mechanisms for efficient message passing with copy avoidance in a distributed system using advanced network devices
US20110134930A1 (en) Packet-based networking system
US20040249948A1 (en) Performing application layer transactions during the connection establishment phase of connection-oriented protocols
Fletcher et al. Mechanisms for a reliable timer-based protocol
JPH11143845A (ja) ネットワークノード間のメッセージ送信用システム及び方法
US20100262859A1 (en) System and method for fault tolerant tcp offload
EP2513794A1 (de) Transparente wiederherstellung von transportverbindungen mit paketübersetzungsverfahren
Farber et al. Thinwire protocol for connecting personal computers to the Internet
CN109067506A (zh) 一种基于多滑动窗口并发的轻量级异步消息实现方法
US7502324B1 (en) TCP retransmission and exception processing in high speed, low memory hardware devices
US7672239B1 (en) System and method for conducting fast offloading of a connection onto a network interface card
Ayuso et al. FT-FW: efficient connection failover in cluster-based stateful firewalls
Paris et al. A high performance erlang TCP/IP stack
Paris et al. Developing a functional TCP/IP stack oriented towards TCP connection replication
Zagorodnov Engineering fault-tolerant TCP/ip services
CN114844826A (zh) 在网络的节点之间的异步套接字复制
Farber et al. RFC0914: Thinwire protocol for connecting personal computers to the Internet

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

17P Request for examination filed

Effective date: 20051007

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LI LU MC NL PT RO SE SI SK TR

AX Request for extension of the european patent

Extension state: AL LT LV MK

RIN1 Information on inventor provided before grant (corrected)

Inventor name: TRACEY, JOHN

Inventor name: SAHU, SAMBIT

Inventor name: PRADHAN, PRASHANT

Inventor name: NAHUM, ERICH

Inventor name: MRAZ, RONALD

Inventor name: FREIMUTH, DOUGLAS

RIN1 Information on inventor provided before grant (corrected)

Inventor name: TRACEY, JOHN

Inventor name: SAHU, SAMBIT

Inventor name: PRADHAN, PRASHANT

Inventor name: NAHUM, ERICH

Inventor name: MRAZ, RONALD

Inventor name: FREIMUTH, DOUGLAS

RIN1 Information on inventor provided before grant (corrected)

Inventor name: TRACEY, JOHN

Inventor name: SAHU, SAMBIT

Inventor name: PRADHAN, PRASHANT

Inventor name: NAHUM, ERICH

Inventor name: MRAZ, RONALD

Inventor name: FREIMUTH, DOUGLAS

DAX Request for extension of the european patent (deleted)
A4 Supplementary search report drawn up and despatched

Effective date: 20090211

RIC1 Information provided on ipc code assigned before grant

Ipc: H04L 29/14 20060101ALI20090205BHEP

Ipc: H04L 29/06 20060101ALI20090205BHEP

Ipc: G06F 11/00 20060101AFI20041115BHEP

17Q First examination report despatched

Effective date: 20090618

GRAP Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOSNIGR1

GRAS Grant fee paid

Free format text: ORIGINAL CODE: EPIDOSNIGR3

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

AK Designated contracting states

Kind code of ref document: B1

Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LI LU MC NL PT RO SE SI SK TR

REG Reference to a national code

Ref country code: CH

Ref legal event code: NV

Representative=s name: IBM RESEARCH GMBH ZURICH RESEARCH LABORATORY INTEL

Ref country code: GB

Ref legal event code: FG4D

Ref country code: CH

Ref legal event code: EP

REG Reference to a national code

Ref country code: DE

Ref legal event code: R084

Ref document number: 60338310

Country of ref document: DE

REG Reference to a national code

Ref country code: IE

Ref legal event code: FG4D

REG Reference to a national code

Ref country code: DE

Ref legal event code: R096

Ref document number: 60338310

Country of ref document: DE

Effective date: 20111103

REG Reference to a national code

Ref country code: GB

Ref legal event code: 746

Effective date: 20111101

REG Reference to a national code

Ref country code: DE

Ref legal event code: R084

Ref document number: 60338310

Country of ref document: DE

Effective date: 20110922

REG Reference to a national code

Ref country code: NL

Ref legal event code: VDEP

Effective date: 20110831

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: FI

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20110831

Ref country code: NL

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20110831

Ref country code: SE

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20110831

REG Reference to a national code

Ref country code: AT

Ref legal event code: MK05

Ref document number: 522865

Country of ref document: AT

Kind code of ref document: T

Effective date: 20110831

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: CY

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20110831

Ref country code: GR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20111201

Ref country code: AT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20110831

Ref country code: SI

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20110831

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: BE

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20110831

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: CZ

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20110831

Ref country code: SK

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20110831

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: EE

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20110831

Ref country code: IT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20110831

Ref country code: PT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20120102

Ref country code: RO

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20110831

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: DK

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20110831

PLBE No opposition filed within time limit

Free format text: ORIGINAL CODE: 0009261

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT

26N No opposition filed

Effective date: 20120601

REG Reference to a national code

Ref country code: DE

Ref legal event code: R097

Ref document number: 60338310

Country of ref document: DE

Effective date: 20120601

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: FR

Payment date: 20120731

Year of fee payment: 10

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: MC

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20120731

REG Reference to a national code

Ref country code: CH

Ref legal event code: PL

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: LI

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20120731

Ref country code: ES

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20111211

Ref country code: CH

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20120731

REG Reference to a national code

Ref country code: IE

Ref legal event code: MM4A

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: BG

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20111130

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: IE

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20120721

REG Reference to a national code

Ref country code: FR

Ref legal event code: ST

Effective date: 20140331

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: TR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20110831

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: LU

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20120721

Ref country code: FR

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20130731

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: HU

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20030721

REG Reference to a national code

Ref country code: DE

Ref legal event code: R082

Ref document number: 60338310

Country of ref document: DE

Representative=s name: KUISMA, SIRPA, FI

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: GB

Payment date: 20220726

Year of fee payment: 20

Ref country code: DE

Payment date: 20220831

Year of fee payment: 20

P01 Opt-out of the competence of the unified patent court (upc) registered

Effective date: 20230423

REG Reference to a national code

Ref country code: DE

Ref legal event code: R071

Ref document number: 60338310

Country of ref document: DE

REG Reference to a national code

Ref country code: GB

Ref legal event code: PE20

Expiry date: 20230720

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: GB

Free format text: LAPSE BECAUSE OF EXPIRATION OF PROTECTION

Effective date: 20230720